Sign in with
Sign up | Sign in
Your question

WD1200JB article; discussion please

Last response: in Storage
Share
March 5, 2002 2:29:40 PM

Here we go, what do you think?

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
Don't step in the sarcasm!
Anonymous
a b G Storage
March 5, 2002 3:41:26 PM

Has anybody tried benchmarking HDs using Norton Ghost 2001/2002. There is nothing interfering, IE NO OS other than DOS, no OS caching, etc. I get a wide variety of numbers with different mobos and HD combos. The new ATA133 drive from Maxtor (I used a 40Gb 7200 ATA133)with an AOpen AK77 Pro 133 mobo produced the highest numbers I've seen. I did a drive to drive copy and got 1700MB/m thru put.
March 5, 2002 5:12:09 PM

28MB/s isn't very impressive, though.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
Don't step in the sarcasm!
Related resources
March 5, 2002 5:31:20 PM

I guess you pay for the performance oO(kinda like a SCSI drive, huh?). The bench marks are really attractive!

To tell you the truth, I didn't like the way they portrayed the ATA133. ATA133 came out because they needed something to get pass the 127GB limitation. Does it really matter if it performs, as long as you can get pass the 28bit LBA limitation. What were they suppose to do to get pass the limitation? Why doesn't ATA100 get the slack? I mean ATA133 at least has the potential to fill the 32bit 33MHz PCI bus (133MB/sec). ATA100 doesn't, right? Just doesn't make since of why they are making a stink about something that really isn't that important. They even use that for the introduction.
**It's probably all marketing anyways.

<font color=red>People and hard drives are like bandwagon fans and sports!</font color=red>
March 5, 2002 11:36:04 PM

From what I understand, ATA 133 has nothing to do with the 127gb limitation. Only a workaround was included in the ATA 133 controller. The same should be able to be done with a ata 100 controller. as the fastest drives now transfer at close to 50 MB/s, the ata 100 controler is far from being saturated. Its marketing plain and simple. And yes an ata 100 controller can saturate the pci bus if it is a dual controller, which is what your going to need to use with either the ata133 or 100 to saturate it.

It's not what they tell you, its what they don't tell you!
March 5, 2002 11:38:53 PM

Quote:
Here we go, what do you think?

Nice drive, and its about time. I really wish the article got more in depth into the different ways the drive cache alogrithm can/is optimized. I really think this is the reason why some IDE drives perform much better than others when setup in an array.

It's not what they tell you, its what they don't tell you!<P ID="edit"><FONT SIZE=-1><EM>Edited by ncogneto on 03/05/02 08:39 PM.</EM></FONT></P>
March 6, 2002 12:01:24 AM

I think we both understand something different! Why would Maxtor come out with a 160GB hard drive to put on a ATA100 controller that only supports 127GB. If you have noticed the ATA133 controller is the only controller that supports 127 + Gigabyte?

<font color=red>People and hard drives are like bandwagon fans and sports!</font color=red>
March 6, 2002 12:12:04 AM

ATA 100 is 28 bit? and ATA 133 is 48 bit right?

<font color=red>:</font color=red> <font color=white>:</font color=white> <font color=blue>:</font color=blue>
March 6, 2002 12:23:21 AM

Correct. If the math is done the 28bit is 127GB and the 48bit is 144 petabytes.

<font color=red>People and hard drives are like bandwagon fans and sports!</font color=red>
March 6, 2002 12:34:56 AM

What I am saying is the controller does not need to be ata 133 to support larger drives. However, you are correct in the fact that the only controller that does at this time is an ata 133 controller. Your dealing with two seperate entities, 1) the ability to transfer at a peak rate of 133 MB/s, and 2) the ability to use drives above 127 gigabytes. Maxtor incorporated both into there ata 133 controller. That does not mean that you could not make a ATA 100 controller capable of useing drives larger than 127 gigabytes.
From Tom's reveiw:

Quote:
Indeed there's not much that speaks in favor of the new standard. Even the latest hard drives do not deliver more than 50 MB/s, and the limit at 127 GB due to limited addressing possibilities has nothing to to with the interface.


It's not what they tell you, its what they don't tell you!<P ID="edit"><FONT SIZE=-1><EM>Edited by ncogneto on 03/05/02 09:36 PM.</EM></FONT></P>
March 6, 2002 11:09:10 AM

Exactly, so why make an arguement about it?

<font color=red>People and hard drives are like bandwagon fans and sports!</font color=red>
March 6, 2002 12:39:39 PM

A somewhat decent article that yet again proves THG staff isn't exactly the best, but they're not all bad either.

My biggest complaint is with the title of the article and the introduction. The proper title of the article is, "Western Digital WD1200JB With 8 MB Cache:
Outperforms SCSI Drives".

Even in the introduction Patrick Schmid states, "Will this be enough to compete with the fastest SCSI drives as well?" Yet do they even provide a single SCSI drive in the benchmarks? Do they even mention things like a 15000rpm SCSI drive with an equal 8MB of cache? How can you title an article like that and drum up interest in SCSI through statements in the introduction of the article and then proceed to have the entire article be completely about IDE?

I've said it before and I'll say it again, THG lacks professionalism. They're not bad, and their benchmarks are useful, but their articles could be written a lot better, and done so with very little additional effort.

Hell, if they even just hired a technical editor who would read and review all articles to suggest improvements or make changes before they publish them, it would make a world of difference in the professionalism of the site.

<pre><font color=green>//error-proof coding</font color=green>
<font color=blue>void</font color=blue> main(){<font color=blue>return</font color=blue>;}</pre><p>
Anonymous
a b G Storage
March 6, 2002 6:48:24 PM

Well, I just buyed the 120 gig mothership. All i can say It's pretty damn fast. It's the 8 mb thats does it.
On the retail box stands : 'Performance rivals 15,000 rpm scsi drives' Well that's a little too much to say. The cpu usage is off course way behind. And the data transfer rate cannot compete. But it's that amount of gig's and those benchmarks It's well worth the price...

and of course the benchmark is based on other tests from scsi drives to compare from that. You can use of course comparing data from other tests, and yes they could have linked you there, but.. And other to compare the benchmark results speaks for themselfs. They compared mostly off course with other IDE disks...

And this one is the best all the way!
March 6, 2002 7:52:26 PM

I wanta 120 gig mothership too.

<font color=red>:</font color=red> <font color=white>:</font color=white> <font color=blue>:</font color=blue>
March 6, 2002 7:56:40 PM

I don't debate that it's a good hard drive, especially for IDE standards. And I don't have any problems with comparing it to other IDE hard drives, since it is one itself.

However, it seems to me that if you claim in the title of the article itself that it "Outperforms SCSI Drives", then you ought to be able to back up that statement in at least some way. And backing up this statement is exactly what Patrick did <i>not</i> do.

Considering the rpm difference itself between top model IDE and top model SCSI, the seek time difference must be noticably different. For that matter, so must the speed at which they read and write. And we all know that the theoretical data throughput of SCSI is considerably more advanced than IDE. So it seems to me that even without any benchmarks to back it up, it's a fair statement to say that the top model SCSI drive would kick the platters off of <i>any</i> IDE drive under <i>any</i> benchmark. (With the possible exception of a cache burst in which they might tie.)

Yet Patrick Schmid seems to have a contrary opinion according to his article, and yet he offers nothing to serve to back this opinion up. I'd call that pretty unprofessional.

If I came onto this forum and said that I've overclocked a Pentium 3 and it'll beat everyone's computers in any benchmark, I'd be severely laughed at, if not flamed. Especially if I then proceeded to <i>not</i> offer <i>any</i> evidence to back up my claim.

So how come THG staff can get away with writing articles that do this kind of thing quite often? I just call it plainly and simply: UNPROFESSIONAL

<pre><font color=green>//error-proof coding</font color=green>
<font color=blue>void</font color=blue> main(){<font color=blue>return</font color=blue>;}</pre><p>
March 6, 2002 9:36:58 PM

slvr_phoenix

You have nailed it right on the head. THG, while still usefull, is slipping. These are not the same quality reviews once provided by the likes of Dr. Thomas Pabst himself.

It's not what they tell you, its what they don't tell you!
March 6, 2002 9:44:28 PM

I am not trying to be argumentive, only correcting an error. I am bringing this up so that some people might not be mislead into assuming that a ata 133 drive would automatically be faster than a ata 100 drive. For this simply is not the matter as this review shows us. It is my opinion that at the present time, ATA 133 is more marketing than substance. Granted the maxtor ata 133 controller has a nice feature ( and btw I highly recomend maxtor drives), but this not due to the fact of its higher alledged max transfer rate.

It's not what they tell you, its what they don't tell you!
March 7, 2002 12:17:41 AM

Agreed! I understand totally. Everything is marketing these days anyway. :wink:

<font color=red>People and hard drives are like bandwagon fans and sports!</font color=red>
March 7, 2002 3:07:50 PM

So true. Even when Dr. Pabst has made some controversial claims, they still had enough evidence behind them that you could debate them from either side. However some of the other writers (Patrick Schmid and Frank Völkel specifically come to mind) have a bad habbit of making some completely unsupportable claims or giving some wild narrow-minded opinions, and doing so fairly regularly. It has gotten to the point where unless it has Dr. Thomas Pabst's name on it, I completely ignore everything but the benchmarks themselves.

What gets me though is just how hard would it be to hire (or 'promote') a single editor to review articles before they're published? Heck, I bet there are plenty of people here who would be willing to do it for free just to get their name on something to put down on a resume. (Or even just for the fun of it.)

It might add a whole hour to the publishing process, and yet it would raise the bar of professionalism of the site back up to good standards.

I mean I can understand some of the typos and bad grammar being due to language translation or English not being everyone's primary language. And I have no problems with that. I mean most people on the planet don't even speak English as a primary language, if at all. I know that the <i>only</i> second verbal language (I won't count computer programming languages) that I know, I can still only barely speak or write as I only spent a year in high school to learn it. So I could hardly judge others for not having perfect skill in their second language.

However, I would find it very difficult to believe that making a claim that an IDE hard drive outperforms a SCSI drive is due to just a language mistranslation. And that's the kind of stuff that makes me hesitate on even reading THG's articles some times.

But hey, it isn't my company, so it isn't my job to fix it. If Dr. Pabst wants his good name being drug through the mud by some of his employees, that's his choice. And I'll just continue to complain every so often in the hopes that someone on high finally listens.

<pre><font color=green>//error-proof coding</font color=green>
<font color=blue>void</font color=blue> main(){<font color=blue>return</font color=blue>;}</pre><p>
March 7, 2002 3:15:31 PM

Was there ever a time when it wasn't? ;) 

<pre><font color=green>//error-proof coding</font color=green>
<font color=blue>void</font color=blue> main(){<font color=blue>return</font color=blue>;}</pre><p>
!